INVESTIGATING POLITICAL-DYNAMICS USING FRACTIONAL-INTEGRATION METHODS

Theory: Many questions central to political science, such as the issue of stability and change in the United States party system, revolve ar ound the degree of persistence or memory in a political process. Fract ional integration techniques, which allow researchers to investigate d ynamic behavior that falls between the stationary and integrated alter natives, provide more precise ways to test hypotheses about the degree of persistence than current modeling strategies. Hypotheses: Choices about the treatment of the time series properties of the data and mode l specification may influence the substantive conclusions drawn about the dynamics of important political processes. Methods: Fractional int egration methods are discussed and compared with common univariate dia gnostic tests. A transfer function model of macropartisanship using fr actional integration techniques is contrasted with traditional ARMA an d ARIMA methods. Results: Fractional integration techniques offer a mo re flexible way to model a time series. Using fractional integration t echniques, we find that macropartisanship is dominated by a strong per manent component, but also contains transitory dynamics in response to changes in economic evaluations and a measure of presidential approva l. Our empirical work shows the importance of taking seriously the tim e series properties of data to ensure valid inferences about the dynam ics of political processes.